Abstarct: Salt domes are most important Geological events that identifying and determining the exact location of them in seismic reflection data is an important factor for characterizing hydrocarbon reservoirs. Boundaries and sub salt are often a poor reflection on seismic data and they are rarly detectable. Because of the nature of seismic data, identifying salt domes and determine the exact location of the seismic data is very time consuming and difficult process. Seismic attributes are tools that will reveal hidden information in seismic data and are able to provide quantitative interpretation of the seismic interpreter. In this thesis, the Histogram of Oriented Gradient (HOG) propose as a seismic attribute to identify and locate fault and salt domes in reflection seismic data. The main objective of the HOG algorithm is extract the desired characteristics of the data that used to detect various objects. HOG method applied on seismic data includes fault and salt dome and with using statistical features that introduce as seismic attributes, distinguish faults and salt dome texture from other textures. Thus, an accurate picture will achieves from faults and salt dome that makes it easy to seismic interpretation. Review the achieved results and comparing these attribute with common texture attributes show that the HOG attribute can be used as an appropriate attributes for identifying salt domes in reflection seismic data.